Gene delivery is a promising method for the treatment of acquired and inherited diseases. A number of viral based systems for gene transfer purposes have been described, such as retroviral systems which are currently the most widely used viral vector systems for this purpose. For descriptions of various retroviral systems, see, e.g., U.S. Pat. No. 5,219,740; Miller and Rosman (1989) BioTechniques 7:980-990; Miller, A. D. (1990) Human Gene Therapy 1:5-14; Scarpa et al. (1991) Virology 180:849-852; Burns et al. (1993) Proc. Natl. Acad. Sci. USA 90:8033-8037; and Boris-Lawrie and Temin (1993) Cur. Opin. Genet. Develop. 3:102-109.
Adeno-associated virus (AAV) systems have also been used for gene delivery. AAV is a helper-dependent DNA parvovirus which belongs to the genus Dependovirus. AAV requires infection with an unrelated helper virus, either adenovirus, a herpesvirus or vaccinia, in order for a productive infection to occur. The helper virus supplies accessory functions that are necessary for most steps in AAV replication. In the absence of such infection, AAV establishes a latent state by insertion of its genome into a host cell chromosome. Subsequent infection by a helper virus rescues the integrated copy which can then replicate to produce infectious viral progeny. AAV has a wide host range and is able to replicate in cells from any species so long as there is also a successful infection of such cells with a suitable helper virus. Thus, for example, human AAV will replicate in canine cells co-infected with a canine adenovirus. AAV has not been associated with any human or animal disease and does not appear to alter the biological properties of the host cell upon integration. For a review of AAV, see, e.g., Berns and Bohenzky (1987) Advances in Virus Research (Academic Press, Inc.) 32:243-307.
The AAV genome is composed of a linear, single-stranded DNA molecule which contains 4681 bases (Berns and Bohenzky, supra). The genome includes inverted terminal repeats (ITRs) at each end which function in cis as origins of DNA replication and as packaging signals for the virus. The ITRs are approximately 145 bp in length. The internal nonrepeated portion of the genome includes two large open reading frames, known as the AAV rep and cap regions, respectively. These regions code for the viral proteins involved in replication and packaging of the virion. In particular, a family of at least four viral proteins are synthesized from the AAV rep region, Rep 78, Rep 68, Rep 52 and Rep 40, named according to their apparent molecular weight. The AAV cap region encodes at least three proteins, VP1, VP2 and VP3. For a detailed description of the AAV genome, see, e.g., Muzyczka, N. (1992) Current Topics in Microbiol. and Immunol. 158:97-129.
The construction of recombinant AAV virions has been described. See, e.g., U.S. Pat. Nos. 5,173,414 and 5,139,941; International Publication Numbers WO 92/01070 (published Jan. 23, 1992) and WO 93/03769 (published Mar. 4, 1993); Lebkowski et al. (1988) Molec. Cell. Biol. 8:3988-3996; Vincent et al. (1990) Vaccines 90 (Cold Spring Harbor Laboratory Press); Carter, B. J. (1992) Current Opinion in Biotechnology 3:533-539; Muzyczka, N. (1992) Current Topics in Microbiol. and Immunol. 158:97-129; and Kotin, R. M. (1994) Human Gene Therapy 5:793-801.
Contemporary recombinant AAV (rAAV) virion production involves co-transfection of a host cell with an AAV vector plasmid and a construct which provides AAV helper functions to complement functions missing from the AAV vector plasmid. In this manner, the host cell is capable of expressing the AAV proteins necessary for AAV replication and packaging. The host cell is then infected with a helper virus to provide accessory functions. The helper virus is generally an infectious adenovirus (type 2 or 5), or herpesvirus.
AAV helper functions can be provided via an AAV helper plasmid that includes the AAV rep and/or cap coding regions but which lacks the AAV ITRs. Accordingly, the helper plasmid can neither replicate nor package itself. A number of vectors that contain the rep coding region are known, including those vectors described in U.S. Pat. No. 5,139,941, having ATCC Accession Numbers 53222, 53223, 53224, 53225 and 53226. Similarly, methods of obtaining vectors containing the HHV-6 homologue of AAV rep are described in Thomson et al. (1994) Virology 204:304-311. A number of vectors containing the cap coding region have also been described, including those vectors described in U.S. Pat. No. 5,139,941.
AAV vector plasmids can be engineered to contain a functionally relevant nucleotide sequence of interest (e.g., a selected gene, antisense nucleic acid molecule, ribozyme, or the like) that is flanked by AAV ITRs which provide for AAV replication and packaging functions. After an AAV helper plasmid and an AAV vector plasmid bearing the nucleotide sequence are introduced into the host cell by transient transfection, the transfected cells can be infected with a helper virus, most typically an adenovirus, which, among other functions, transactivates the AAV promoters present on the helper plasmid that direct the transcription and translation of AAV rep and cap regions. Upon subsequent culture of the host cells, rAAV virions (harboring the nucleotide sequence of interest) and helper virus particles are produced.
When the host cell is harvested and a crude extract is produced, the resulting preparation will contain, among other components, approximately equal numbers of rAAV virion particles and infectious helper virions. rAAV virion particles can be purified away from most of the contaminating helper virus, unassembled viral proteins (from the helper virus and AAV capsid) and host cell proteins using known techniques. Purified rAAV virion preparations that have been produced using infection with adenovirus type-2 contain high levels of contaminants. Particularly, 50% or greater of the total protein obtained in such rAAV virion preparations is free adenovirus fiber protein. Varying amounts of several unidentified adenoviral and host cell proteins are also present. Additionally, significant levels of infectious adenovirus virions are obtained, necessitating heat inactivation. The contaminating infectious adenovirus can be inactivated by heat treatment (56.degree. C. for 1 hour) and rendered undetectable by sensitive adenovirus growth assays (e.g., by cytopathic effect (CPE) in a permissive cell line). However, heat treatment also results in an approximately 50% drop in the titer of functional rAAV virions.
Production of rAAV virions using an infectious helper virus (such as an adenovirus type-2, or a herpesvirus) to supply accessory functions is undesirable for several reasons. AAV vector production methods which employ a helper virus require the use and manipulation of large amounts of high titer infectious helper virus which presents a number of health and safety concerns, particularly in regard to the use of a herpesvirus. Also, concomitant production of helper virus particles in rAAV virion producing cells diverts large amounts of cellular resources away from rAAV virion production, possibly resulting in lower rAAV virion yields.
More particularly, in methods where infection of cells with adenovirus type-2 are used to provide the accessory functions, more than 95% of the contaminants found in the purified rAAV virion preparations are derived from adenovirus. The major contaminant, free adenovirus fiber protein, tends to co-purify with rAAV virions on CsCl density gradients due to a non-covalent association between the protein and rAAV virions. This association makes separation of the two especially difficult, lowering rAAV virion purification efficiency. Such contaminants may be particularly problematic since many adenoviral proteins, including the fiber protein, have been shown to be cytotoxic (usually at high concentrations), and thus may adversely affect or kill target cells. Thus, a method of producing rAAV virions without the use of infectious helper viruses to provide necessary accessory functions would be advantageous.
A number of researchers have investigated the genetic basis of accessory functions, particularly adenovirus-derived functions. Generally, two approaches have been used to attempt to identify those adenoviral genes that are involved in AAV replication: examination of the ability of various adenovirus mutants to provide accessory functions; and the study of the effect of transfected adenoviral genes or regions on AAV replication in the absence of adenovirus infection.
Studies with various adenovirus mutants that are capable of supporting AAV replication (e.g., by supplying necessary accessory functions) at or about the levels obtained by infection with a wild-type adenovirus have demonstrated that particular adenovirus genes or gene regions are not involved in AAV replication. However, loss-of-function data from such studies have failed to provide conclusive information that a particular gene region is involved with AAV replication since many of the adenovirus genes and control regions are overlapping and/or incompletely mapped.
Particularly, adenovirus mutants with fairly well characterized mutations in the following genes or gene regions have been tested for their ability to provide accessory functions necessary for AAV viral replication: E1a (Laughlin et al. (1982) J. Virol. 41:868, Janik et al. (1981) Proc. Natl. Acad. Sci. USA 78:1925); E1b (Laughlin et al. (1982), supra, Janik et al. (1981), supra, Ostrove et al. (1980) Virology 104:502); E2a (Handa et al. (1975) J. Gen. Virol. 29:239, Straus et al. (1976) J. Virol. 17:140, Myers et al. (1980) J. Virol. 35:665, Jay et al. (1981) Proc. Natl. Acad. Sci. USA 78:2927, Myers et al. (1981) J. Biol. Chem. 256:567); E2b (Carter, B. J. (1990) "Adeno-Associated Virus Helper Functions," in CRC Handbook of Parvoviruses, vol. I (P. Tijssen, ed.); E3 (Carter et al. (1983) Virology 126:505); and E4 (Carter et al. (1983), supra, Carter, B. J. (1995), supra). Poorly characterized adenovirus mutants that were incapable of DNA replication and late gene synthesis have also been tested (Ito et al. (1970) J. Gen. Virol. 9:243, Ishibashi et al. (1971) Virology 45:317).
Adenovirus mutants with defects in the E2b and E3 regions have been shown to support AAV replication, indicating that the E2b and E3 regions are probably not involved in providing accessory functions (Carter et al. (1983), supra). Mutant adenoviruses defective in the E1a region, or having a deleted E4 region, are unable to support AAV replication, indicating that the E1a and E4 regions are likely required for AAV replication, either directly or indirectly (Laughlin et al. (1982), supra, Janik et al. (1981), supra, Carter et al. (1983), supra). Studies with E1b and E2a mutants have produced conflicting results. Further, adenovirus mutants incapable of DNA replication and late gene synthesis have been shown to be permissive for AAV replication (Ito et al. (1970), supra, Ishibashi et al (1971), supra). These results indicate that neither adenoviral DNA replication nor adenoviral late genes are required for AAV replication.
Transfection studies with selected adenoviral genes have been used in an attempt to establish whether a transfected set of adenovirus genes is capable of providing the same level of accessory functions for AAV replication as that provided by an adenovirus infection. Particularly, in vitro AAV replication has been assessed using human 293 cells transiently transfected with various combinations of adenovirus restriction fragments encoding single adenovirus genes or groups of genes (Janik et al. (1981), supra). Since the above-described transfection studies were done in cells that stably express the adenovirus E1a and E1b regions, the requirement for those regions could not be tested. However, it was deduced that the combination of three adenoviral gene regions, VA I RNA, E2a and E4, could provide accessory functions (e.g., support AAV replication) at a level that was substantially above background, but that was still approximately 8,000 fold below the level provided by infection with adenovirus. When all combinations of two of the three genes were tested, the accessory function levels ranged between 10,000 to 100,000 fold below the levels provided by infection with adenovirus.
Transfection studies with selected herpes simplex virus type-1 (HSV-1) genes have also been conducted in an attempt to establish whether a transfected set of HSV-1 genes is capable of providing the same level of accessory functions for AAV replication as that provided by an HSV-1 infection. Weindler et al. (1991) J. Virol. 65:2476-2483. However, such studies were limited to identifying only those HSV-1 genes necessary to support wild-type AAV replication, not rAAV production. Further, the identified HSV-1 accessory functions were significantly less efficient at supporting AAV replication than adenovirus-derived functions.
Accordingly, there remains a need in the art to identify a subset of the adenovirus genome or functional homologues of the adenovirus genome, that include only those accessory functions required for rAAV virion production. The subset can then be included in an accessory function vector or system which, when introduced into a suitable host cell, supports the production of rAAV virions in an amount that is substantially equivalent to, or greater than, the amount produced using an adenovirus infection. Further, there remains a need to provide an accessory function system that is capable of producing commercially significant levels of rAAV virion particles without also generating significant levels of infectious adenovirus virions, or other contaminating by-products.